BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.coscup.org//coscup-2026//speaker//XTRQDX
BEGIN:VTIMEZONE
TZID:CST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:CST
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-coscup-2026-LSRELU@pretalx.coscup.org
DTSTART;TZID=CST:20260809T101000
DTEND;TZID=CST:20260809T104000
DESCRIPTION:scx_lavd is a sched_ext scheduler implemented in BPF — dynami
 cally loadable at runtime without a kernel rebuild. It started in 2024 as 
 a Linux gaming experiment for Valve's Steam Deck\, and has since matured i
 nto a serious contender for server workloads. At Linux Plumbers Conference
  2025\, Meta engineers presented LAVD as a candidate default scheduler for
  their server fleet. This session would reveal its core design ideas and h
 ow recent development benefits cloud and server workloads — such as the 
 `cpu.max` cgroup controller\, IRQ-aware task steering\, latency criticalit
 y propagation\, and the task-size-aware load balancer. Beyond cloud\, the 
 same design supports power-aware mobile/edge deployments and hybrid CPU ar
 chitectures — making LAVD relevant across cloud\, automotive\, and mobil
 e vendors.
DTSTAMP:20260713T132600Z
LOCATION:TR213
SUMMARY:LAVD: From Gaming to Server Fleets — a BPF Scheduler for Cloud\, 
 ML\, and Edge - Gavin Guo
URL:https://pretalx.coscup.org/coscup-2026/talk/LSRELU/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-coscup-2026-ZESBXN@pretalx.coscup.org
DTSTART;TZID=CST:20260809T130000
DTEND;TZID=CST:20260809T133000
DESCRIPTION:People who try to learn English as a second language usually qu
 it their flashcard app within two weeks. The hard part isn't memorising de
 finitions. It's that a new word only sticks after a dozen real-world encou
 nters\, and a flashcard never gives you that. We introduce the concept "Am
 bient Learning" — learning from natural conversation.\n\nCOCA MCP is an 
 useful tool to suggest the vocabularies chat to an LLM. The server sits be
 hind the Corpus of Contemporary American English — 60\,000 words ranked 
 by frequency — and exposes a small set of MCP tools. When a client like 
 Claude is talking to a learner\, it asks the server which words are due\, 
 and gets back a short list scheduled by FSRS-4.5. The model uses those wor
 ds in its reply. It then tells the server which ones it actually used\; th
 e server records the exposure and reschedules.\n\nThe session walks throug
 h a modern English learning methodology on the running site at coca.gaving
 uo.cc — how to use MCP technology to improve the English ability.
DTSTAMP:20260713T132600Z
LOCATION:AU
SUMMARY:COCA MCP: Learning English While Chatting with an LLM - Gavin Guo
URL:https://pretalx.coscup.org/coscup-2026/talk/ZESBXN/
END:VEVENT
END:VCALENDAR